@bakpakin https://github.com/janet-lang/janet-lang.org/blob/master/content/docs/index.mdz#L21-L26 I'd send in a PR for this part of the website to update the docs to indicate how to download just the C++ build tools?
Yeah, I mentioned Visual Studio because it seemed more straight forward but the build tools are easy and smaller
when i view from ff i can reproduce.
all of the links along the left side suffer from the same problem when viewed in ff for me, whereas not so from brave.
just tried with chromium and that also has broken links.
I can reproduce it across Safari and Brave on macOS. The problem appears to be caused by the fact there are in fact two 'Introduction' files in the website directory (one is at
/introduction.html and one is at /docs/index.html). You normally wouldn't notice because the website itself doesn't have a link to the /introduction.html but that file is linked from the Janet GitHub repo README.
I think the reason two files are being generated is because the frontmatter to the 'Introduction' includes a
:permalinks key. The other files in the content/docs/ directory don't have that key based on a cursory look. My guess is that you can solve the problem by updating the link in the README and removing the :permalinks key in content/docs/index.mdz.
There were some perma links added but the issue likely with relative anchor tags not working in the new permalink s
I haven't really fixed the issue but I have been trying to fit rid of the old linka
for mendoza
The watch functionality is current very simepl
@bakpakin:matrix.org
I seem to understand the soul of functional programming languages, that is, lists, and all operations are list-oriented. A value is also a list, and an empty value can also be a list. All functions can be designed to process lists, and even grammatical structures can be written as lists
That is maybe the soul of one branch of functional languages, I don't think it speaks for all of them.
Then things like first class functions / composition
HOF
many languages that aren't functional can do list-oriented programming - I like it as it can remove a large class of bugs/checks - instead of checking if a single thing is null/nil, or 1, or many, just write a function that works on a collection, then the logic for the function can just map the collection whether its 0 items (null/nil), 1 item, or many items
prolog and erlang are both very good functional languages that are pretty similar to your description of list-oriented
even "strings" in them are just arrays of integers corresponding to the ascii codes
Erlang is from '85 when they didn't have the efficient persistent data structures that e.g. Clojure uses
So they said, lists is all you get. We can make guarantees for lists.
I think Erlang only got hashmaps in the late 2010s (might be off on that)
I've heard people say with a straight face that the main tenant of functional programming is static typing ala Haskell
Whereas the main functional languages in my mind are Clojure and Erlang, both dynamically typed... so the word "functional" doesn't seem to mean much ;)
I think it's pretty obvious that the main tenet of functional programming is functions, but hey, people have their opinions.
Haha, that's true - I think when comparing functional/procedural/oop/logical languages, that goes without saying though - since the vast majority allow use of functions, if you take "applying functions" to mean, simply calling a function and doing something with the result - its difficult, to impossible to have the standard functional language "feel" in a language that doesn't allow first class functions/HOF
A good sample of a language that allows functions, but has no HOF/FCF would be Godot Script - you can very clunkily implement something like a generic 'map' using it's call() to a stringly function input, but trying to implement a compose() from scratch in it is very difficult to impossible
It's kind of a moving target, right? Backus coined the term (or a term) in '77 - when first class functions weren't really a gleam in most languages' eyes. There's also a difference between what you might say is a functional language - any language that has first-class functions - and functional style - the styles of programming that have evolved over many years within those languages, that might require first class functions but are not identical with them.
@bakpakin:matrix.org A little late to answer, but "the main tenet of functional programming is functions" is pretty meaningless ;) Every language since.. I don't know, Fortran? has functions. That's why I don't like the term "functional language" or "functional programming." Heck, every C programmer spends all day writing functions. I guess the confusion stems from the fact that what most imperative languages call functions should technically be called "subroutines" since they're not functions in the mathematical sense. But that boat sailed 50 years ago.
It seems to me that "expressions" vs "statements" is a much better dividing line. In "functional programming" you typically arrange functions in a way that yield a result ("result flows up and to the left"). In imperative languages you make "statements" that tell the turing machine what to do, always manipulating some kind of state at some specific location. Control flows down and to the right.
But saying "expressive programming" is probably also completely meaningless and won't mean anything to anyone. "I express my thoughts in Java!"
That was mostly a tongue-in-cheek answer, but the point was that if everyone brings their own definitions, the word looses it's original meaning. Which most certainly was programming using constructs that behaved as mathematical functions.